Range of Motion

Under the LCM, the ALIF model showed relative stability, compared with the INT model; the ROM was reduced obviously in flexion (-97.7 %), extension (-88.6 %), and lateral bending (-72.9 %), but less in torsion (-57.0 %), compared with the INT model. In contrast, the ADR model had a large ROM increase in extension (+81.1 %) (Figures 4.3(a)), torsion (+67.9 %) (Figure 4.4(a)), and lateral bending (+44.5 %) (Figure 4.5(a)), but less in flexion (+9.2 %) (Figures 4.2(a)). The ROM values of the LCM are listed in Table 3.2.

Under the HCM, the ROM of the ALIF model was reduced obviously in flexion (-83.7 %), extension (-82.4 %), and lateral bending (-69.2 %), but less in torsion (-47.6 %), compared with the INT model. In contrast, the ADR model had a large ROM increase in extension (+45.1 %) (Figures 4.3(b)), torsion (+42.7 %) (Figure 4.4(b)), and lateral bending (+37.1 %) (Figure 4.5(b)), but less in flexion (+6.4 %) (Figures 4.2(b)). The ROM values of the HCM are listed in Table 3.3.

Overall, both control methods can provide similar stability in the ALIF model.

However, in the ADR model, the LCM showed prominently higher ROM increase than the HCM, especially in extension and torsion.

Comparison of the implant level effect between present FE models and previous studies under the HCM are listed in Table 4.1. For the implant level, the ALIF model showed similar stability with both control methods. Oxland and Lund indicated that anterior fusion plus posterior pedicle screw fixation can improve stabilization in all motions [93]. Gerber et al. [94] indicated that anterior cage plus posterior pedicle screw fixation did not provide significant stability in torsion with the LCM. This behavior of the LCM is similar to the in vitro test of Panjabi et al. [95] using the HCM, in which the ROM decreased by 77.4 % in flexion-extension, by 36.4 % in torsion and by 65.7 % in lateral bending. The results of our ALIF model are in agreement with most of the in

vitro test results [93-96], in that the fusion level can provide good stability in flexion, extension, and lateral bending, but is not so in torsion, regardless of whether the LCM or HCM is used.

The implant level of the ADR model shows significantly increased ROM in extension, torsion, and lateral bending under both control methods. In addition, the LCM showed higher ROM than the HCM, especially in extension and torsion. These characteristics of the ADR model are in agreement with a previous report by Goel et al. [71].

Therefore, this study suggests that both control methods can be adopted to evaluate the implant level of the fusion model, and similar stabilizing characteristics can be expected to be found. On the other hand, the effects on the implant level of the ADR increased ROM with the LCM, especially in extension (81.1 % vs. 45.1 %) and torsion (67.9 % vs. 42.7 %). Thus, LCM analysis might indicate a higher risk for patients with ADR implants. The present study suggests that the LCM might emphasize the effect on the implant level of the non-fusion model.

Adjacent level effects (ALEs)

The ALE% of the ROM was defined as the averaged percentage changes of the ROM from whole non-operated levels. Under the LCM, the ALE% of the ALIF model in flexion (+10.3 %) and extension (-7.3 %) were small, and in torsion and lateral bending were even smaller (average within 6 %). The ALE%

values of the ADR model were close to those of the INT model in flexion (Figures 4.2(a)), extension (Figures 4.3(a)), torsion (Figures 4.4(a)), and lateral bending (average within 6 %) (Figures 4.5(a)).

Under the HCM, the ALE% of the ALIF model increased in flexion (+25.6

%), extension (+28.6 %), torsion (+16.7 %), and lateral bending (+22.8 %), as compared with the INT model. In contrast, the ALE% of the ADR model decreased obviously in extension (-13.6 %) (Figures 4.3(b)), torsion (-14.9 %)

(Figures 4.4(b)), and lateral bending (-13.0 %) (Figures 4.5(b)), but less in flexion (-2.9 %) (Figures 4.2(b)), as compared with those with the INT model.

The HCM increased the ALE% in ROM more than when using the LCM in the ALIF model; on the other hand, the HCM decreased the ALE% more than when using the LCM in the ADR model, especially in extension, torsion, and lateral bending.

The ALEs between present FE models and previous studies under the HCM are listed in Table 4.1. For the adjacent levels, the ALIF model shows a significantly increased ALE%, using the HCM. As mentioned previously, conflicting ALE% results were found with the LCM. The ALE% of the HCM determined in the current study are in the range of the values reported in the literature [73, 95-96], which showed significantly increased ALE% (Table 4.1).

However, a few inconsistencies in ALE% were still noticed. In lateral bending, significantly increased ALE% with fusion was reported (average, +20.7 %) [73];

in contrast, a small ALE% with fusion was also found (average, +4.1 %) [95].

This discrepancy was also revealed in torsion [73, 95-96]. These different results might be explained that different fusion techniques were simulated between their studies. Despite these differences in ALEs, this study has shown that the HCM could emphasize the ALEs more than the LCM on the fusion model.

The ALE% of the ADR model was close to the INT model with the LCM, while it was significantly decreased with the HCM. This trend of an ALE%

decrease was also found in other studies using the HCM [71, 95-96].

Overall, this study suggests that the HCM should be more effective in evaluating the ALEs of the fusion model. On the other hand, the HCM may decrease ALEs of the ADR. Verification of the influence of these abnormal motions requires more evidence through clinical research. Therefore, the present study suggests that both control methods should be used in evaluating ALEs of non-fusion models.

Clinical relevance of loading condition

Goel et al. [71] proposed that the patient’s main aim following surgery is to return to normal daily life. Thus, the surgically treated spine should be able to go through the same ROM as in a normal person. However, in real life, people sustain the same external moments during lifting activities whether or not they have had surgery, making the LCM useful for evaluating this condition. The present study suggests that these two analytical methods could be used to predict specific conditions in the patient’s daily life. The HCM is suitable for evaluating the patient’s daily life motion during restoration after surgery, while the LCM is suitable for evaluating the patient’s normal lift work-loading condition after surgery [97].

In summary, the present study indicates the difference in ROM changes between the LCM and HCM after using a fusion or non-fusion implant. Similar trends were found in present FE simulations of fusion or non-fusion implants compared to previous studies. In a way, these data validate the predictions of the ALIF and the ADR FE models. In addition, the present study suggests that these two analytical methods could be used to predict specific conditions in the patient’s daily life.

Table 4.2: The implant and adjacent level effects on the lumbar spine after implantation of an anterior cage or an artificial disc were compared with previous finite element and in vitro studies under the hybrid control method.

Increase in ROM normalized to intact (%) Implant level effect

Authors Fusion or non-fusion

implants Flexion Extension Torsion Lateral Bneding

Non-operated adjacent levels effect (ALE%) SynCage plus

Note: An asterisk (*) denotes the non-fusion spinal implant model.

Flexion-LCM

Changes in ROM (% of Intact) ALIF

Changes in ROM (% of Intact) ALIF

ADR

(b)

Figure 4.2: Changes in the ROM under flexion: (a) LCM results; (b) HCM results.

Extension-LCM

Changes in ROM(% of Intact) ALIF

Changes in ROM(% of Intact) ALIF

ADR

(b)

Figure 4.3: Changes in the ROM under extension: (a) LCM results; (b) HCM results.

Torsion-LCM

Changes in ROM (% of Intact) ALIF

Changes in ROM (% of Intact) ALIF

ADR

(b)

Figure 4.4: Changes in the ROM under torsion: (a) LCM results; (b) HCM results.

Lateral Bending-LCM

Changes in ROM (% of Intact) ALIF

Changes in ROM (% of Intact) ALIF

ADR

(b)

Figure 4.5: Changes in the ROM under lateral bending: (a) LCM results; (b) HCM results.

4.3. Facet Contact Force under Extension and Torsion